Urinary proteomic diagnosis of coronary artery disease: identification and clinical validation in 623 individuals

Abstract

OBJECTIVES: We studied the urinary proteome in a total of 623 individuals with and without coronary artery disease (CAD) in order to characterize multiple biomarkers that enable prediction of the presence of CAD. METHODS: Urine samples were analyzed by capillary electrophoresis coupled online to micro time-of-flight mass spectrometry. RESULTS: We defined a pattern of 238 CAD-specific polypeptides from comparison of 586 spot urine samples from 408 individuals. This pattern identified patients with CAD in a blinded cohort of 138 urine samples (71 patients with CAD and 67 healthy individuals) with high sensitivity and specificity (area under the receiver operator characteristic curve 87%, 95% confidence interval 81-92) and was superior to previously developed 15-marker (area under the receiver operator characteristic curve 68%, P < 0.0001) and 17-marker panels (area under the receiver operator characteristic curve 77%, P < 0.0001). The sequences of the discriminatory polypeptides include fragments of alpha-1-antitrypsin, collagen types 1 and 3, granin-like neuroendocrine peptide precursor, membrane-associated progesterone receptor component 1, sodium/potassium-transporting ATPase gamma chain and fibrinogen-alpha chain. Several biomarkers changed significantly toward the healthy signature following 2-year treatment with irbesartan, whereas short-term treatment with irbesartan did not significantly affect the polypeptide pattern. CONCLUSION: Urinary proteomics identifies CAD with high confidence and might also be useful for monitoring the effects of therapeutic interventions.